Procedure 1- We started this task by importing all the necessary libraries 2 - As first step after uploading the table we have proceeded with the standardization of the columns, turning the management of the information easier for us 3 - To be able to visualize all the information provided on the table, we applied the code to display all the columns of the table 4 - Then we have created a table with only the numeric columns from the previous table, for us to relate all the numeric information trying to find the best correlations 5 - We dediced to find the correlation between the numeric variables and the variable that we defined that would show us best the value of each player. We have appied a filter to show only correlations above 0.8 6 - Next step was creating a heat map, allowing us to check the strongness of the variable correlations and relations 7 - We created several scatter plot with the variables that we found that were best related with the 'ova' variable. It allowed us to see in a graphic way our variable related with the others ? - Then we have tested different models with different variables, to check the error margins that each value would give us ? - Based on these models, we picked the one that would fit best our goal, and applied it on our second table and asked it to predic the 'ova' (overall) of each player on the second table. We have applied our model as sorted the entries of the second table staring with the highes percentage of the columns 'pred_ova' ? - Based on our final results and taking into consideration our budget (100M) we have identified the following players as the best options to our team:
- Neuer
- Ronhaar
- Allan Thank you for taking your time to read this doc. Best regards, Haissa and Joao